Surgical treatment of bone tumours often requires generous resection
of bone, leaving defects which are difficult to span. Freezing tumours
with liquid nitrogen was introduced in the late 1960's as an adjuvant
treatment to extend the surgical margin of excision for intralesional
resection or for curettage by pouring or spraying the nitrogen directly
into the bone cavity [1-9]. Animal trials by Gage et al. 
have shown that devitalised bone matrix can serve as a framework for
new periostal and endostal growth, and hence that the former tumour
space can be bridged with autologous, healthy bone tissue. However, the
freezing procedure is difficult to control, and therefore harbours
risks of injury for the patient [11,12] and the surgical team, as well as of gas embolisms caused by evaporation  and spread of the liquid nitrogen.
Aside from the open use of liquid nitrogen, closed systems for treating bone tumours have also been used , although they never became popular because the cooling power of the cryoprobes then used was low compared to their diameter . Recent technical advances  made it possible for us to develop new probes for cryoablation of bone tissue and test these in animal trials [17,18].
The efficiency of these procedure and the extent of tissue distruction
is well documented in a former study with a comparable setup .
Various complications have been reported, ranging from soft tissue wound infection and fractures  to bone marrow and fat embolism caused by the spread of the ice front due to an increase in intramedullar pressure .
These miniature cryoprobes with a minimised diameter allow precise
control of the freezing process, thus avoiding uncontrolled freezing of
soft parts and healthy bone tissue, as well as a sudden expansion of
the ice front. Aim of this animal trial was to determine whether the
use of modern miniature cryoprobes can avoid the above described